Académique Documents
Professionnel Documents
Culture Documents
Geoderma
journal homepage: www.elsevier.com/locate/geoderma
Dipartimento di gestione dei Sistemi Agroalimentari e Ambientali (DiGeSA), Universit degli Studi di Catania, Via S. Soa, 100, 95123 Catania, Italy
Dipartimento di Scienze Agrarie e Forestali (SAF), Universit degli Studi di Palermo, Viale delle Scienze, 90128 Palermo, Italy
a r t i c l e
i n f o
Article history:
Received 28 February 2014
Received in revised form 18 July 2014
Accepted 27 July 2014
Available online 8 August 2014
Keywords:
BEST (Beerkan Estimation of Soil Transfer
parameters) procedure
Soil water retention
Saturated soil hydraulic conductivity
Simplied Falling Head technique
a b s t r a c t
Establishing the ability of the Beerkan Estimation of Soil Transfer parameters (BEST) procedure to reproduce soil
properties is necessary for specic soil types. In this investigation, the BEST predictions for a sandy loam soil were
compared with water retention data obtained by a standard laboratory method and with the saturated soil hydraulic conductivity, Ks, obtained by both the Wu et al. (1999) method, applied to the BEST inltration data,
and the Simplied Falling Head (SFH) technique. When the original BEST-slope algorithm with the inltration
constants xed at = 1.9 and = 0.79 was applied, the agreement between the predicted and the measured
retention data was satisfactory in terms of similarity of the means and correlation and coincidence between
the regression and identity lines. The prediction of Ks at a sampling point differed by not more than a factor of
two from the Ks value obtained by the Wu et al. (1999) method. The SFH technique yielded Ks values approximately ve times higher than those of BEST, probably because soil disturbance during water application, swelling
and air entrapment phenomena had a lower impact on the measured inltration data with the former technique.
In conclusion, BEST is a promising approach for easily characterizing a soil, but its method of application should
be adapted to the particular situation under consideration. Additional investigations carried out on different soils
would allow development of more general procedures for applying BEST.
2014 Elsevier B.V. All rights reserved.
1. Introduction
The hydraulic characteristic curves, i.e., the relationships between
soil water pressure head, h, volumetric water content, , and hydraulic
conductivity, K, are generally determined with laboratory and eld
methods of differing accuracy and experimental effort. The availability
of different methods should allow researchers to choose the most appropriate technique for interpreting and simulating a particular hydrological process occurring in a given soil. However, there is also the need
to simplify experimental procedures, especially because the economic
resources for soil hydraulic characterization are often scarce.
Lassabatre et al. (2006) proposed the Beerkan Estimation of Soil
Transfer parameters (BEST) procedure to easily and rapidly estimate
the (h) and K() curves. BEST uses an inltration experiment in the
eld with a zero pressure head on a circular soil surface and a few basic
soil physical determinations (particle size distribution (PSD), bulk
density, and initial and nal water content). BEST focuses on the van
Genuchten (1980) relationship for the water retention curve with the
Burdine (1953) condition and the Brooks and Corey (1964) relationship
for hydraulic conductivity. Due to its simplicity and the physical soundness of the employed relationships and procedures, BEST is receiving
Corresponding author.
http://dx.doi.org/10.1016/j.geoderma.2014.07.024
0016-7061/ 2014 Elsevier B.V. All rights reserved.
301
Table 1
Clay, cl, silt, si (250 m), and sand, sa, percentages, dry soil bulk density, b, and organic
carbon content, OC, at the eld experimental site (sample size, N = 32 for cl, si, sa and OC,
and N = 64 for b).
Variable
Mean
cl (%)
si (%)
sa (%)
b (Mg m3)
OC (%)
10.4
19.9
69.7
1.25
1.25
7.6
3.8
5.8
6.2
21.8
using stainless steel cylinders with an inner volume of 104 m3 to determine the soil water retention curve. For each sample, the volumetric
soil water content at 11 pressure heads, h, was determined by a sandbox
(h = 0.01, 0.025, 0.1, 0.32, 0.63, 1.0 m) and a pressure
plate apparatus (h = 3, 10, 30, 60, 150 m). For each sample,
the parameters of the van Genuchten (1980, vG) model for the water retention curve with the Burdine (1953) condition were determined by
tting the following relationship to the data:
"
r
1
s r
m 1
h
hg
!n #m
1a
2
n
1b
where (L3 L3) is the volumetric soil water content, h (L) is the soil
water pressure head, n (N2) and m are shape parameters, and hg (L),
s (L3 L3, eld saturated soil water content), and r (L3 L3, residual
soil water content) are scale parameters. The tting was performed by
an iterative nonlinear regression procedure, which nds the values of
the optimized parameters by minimizing the sum of the squared residuals between the model and the data. This procedure was applied using
the SOLVER routine of Microsoft Excel software (Microsoft Company,
Redmond, WA, USA). According to the BEST procedure, r was set equal
to zero. To evaluate the tting performance of the vG model to the measured water retention data, the relative error, Er (%), was calculated for
each sampling point using the following relationship (Lassabatre et al.,
2006):
v
uX
2
u q
u
m;i vG;i
u
u
Er 100 u i1 q
2
u X
t
m;i
i1
where m,i denotes the experimental data, i.e., the measured soil water
content at a given pressure head, vG,i is the corresponding modeled soil
water content, and q is the number of the (h, ) data pairs. According to
Bagarello and Iovino (2012), Er 5% can be assumed to be indicative of
a satisfactory tting ability of the model. The residuals, i, were also calculated by the following relationship:
i vG;i m;i :
302
Table 2
Considered scenarios to apply the BEST procedure and valid sampling points, i.e., yielding
valid estimates of soil sorptivity and saturated soil hydraulic conductivity (sampled
points = 32).
Progressive Scenario Algorithm Shape parameter
number
ID
estimation
Valid sampling
points, N
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
0.75
0.61
0.7
0.79
0.61
0.7
0.79
0.61
0.7
0.79
0.75
0.61
0.7
0.79
0.61
0.7
0.79
0.61
0.7
0.79
0.75
0.61
0.7
0.79
0.61
0.7
0.79
0.75
0.61
0.7
0.79
0.61
0.7
0.79
32
32
32
32
32
32
32
32
32
32
32
32
32
32
32
32
32
32
32
32
31
28
31
31
28
28
28
31
28
31
31
28
28
28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
28
29
30
31
32
33
34
38
39
40
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Slope
Intercept
Intercept
Intercept
Intercept
Intercept
Intercept
Intercept
Intercept
Intercept
Intercept
Intercept
Intercept
Intercept
Intercept
FIT
FIT
FIT
FIT
FIT
FIT
FIT
FIT
FIT
FIT
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
FIT
FIT
FIT
FIT
FIT
FIT
FIT
MM
MM
MM
MM
MM
MM
MM
0.6
0.1
0.1
0.1
1.0
1.0
1.0
1.9
1.9
1.9
0.6
0.1
0.1
0.1
1.0
1.0
1.0
1.9
1.9
1.9
0.6
0.1
0.1
0.1
1.9
1.9
1.9
0.6
0.1
0.1
0.1
1.9
1.9
1.9
Z0 "
hi
1
h
1
hg
!n !m #
dh
using the tted parameters of the vG model to the data and calculating
the parameter of the Brooks and Corey (1964) hydraulic conductivity
function according to BEST.
The SFH technique (Bagarello et al., 2004) was also applied at the 32
sites. In all cases, the 0.30-m inner diameter ring was inserted 0.12 m
into the soil at a distance of approximately 0.150.20 m from the ring
used for the BEST run. Ring insertion was conducted using a rubber
hammer, ensuring that the upper rim of the ring remained horizontal
during the insertion process. The soil cores collected a few days before
applying the SFH inltration test allowed us to determine the soil
water content at the time of the run. This value and the saturated soil
water content, s, estimated using the measured b and assuming a
soil particle density of 2.65 Mg m3 (Bagarello et al., 2004) were used
to establish the volume of water to be used for the SFH test. In practice,
a common water volume of 2 L was used at each sampling point to ensure one-dimensional ow in all cases. The Ks values were determined
by the measured time, ta (T), from the application of the known volume
of water to the instant at which it had completely inltrated. The choice
of using the SFH technique to obtain independent Ks data was made
because the relative performances of the two techniques (BEST, SFH)
have been tested in other recent investigations (Bagarello et al., 2014a,
b) and this circumstance assisted in the interpretation of the method
comparison. The * parameter calculated with Eq. (4) was also used to
determine Ks with the SFH equation.
The BEST and SFH runs differed by ring insertion depth because
the inltration process has to be three-dimensional in the former
case (Lassabatre et al., 2006) and one-dimensional in the latter
one (Bagarello et al., 2004). However, this last difference was not
considered to represent a limitation of the established comparison
because rings determine downward ow and ring inltrometers are expected to essentially measure vertical soil water transmission parameters (Reynolds and Elrick, 2005). The choice not to describe in detail
the methods and procedures used in this investigation was made for
reasons of brevity and also because an unavoidably synthetic method
description would not be enough to allow a reader to reproduce the
experiment.
To assess the performances of BEST for predicting soil water retention, it was considered that this procedure assumes aprioristically that
the vG model is usable and it does not offer any other alternatives.
Therefore, the assessment of BEST was carried out in terms of the tted
(vG) soil water content values against the corresponding values estimated by BEST (B). For a comparison between the two datasets, the
paired differences, i.e., vG B for a given sampling point and pressure
head, were calculated and the hypothesis of normality of these differences was checked by the Lilliefors (1967) test. Then, a two-tailed paired
t-test was used to compare the means (Helsel and Hirsch, 1992). A linear
regression analysis between B and vG was also carried out, and the statistical signicance of the correlation coefcient, R, was established by a
one-tailed t-test. The normality of the residuals was tested, and the condence intervals for both the intercept and the slope of the linear regression line were calculated. Similar comparisons were also carried out with
reference to the Ks values obtained with BEST, method 2 by Wu et al.
(1999) and the SFH technique. Statistical signicance was assessed at a
P = 0.05 probability level, and the 95% condence intervals for the intercept and the slope were calculated, unless otherwise specied.
303
between vG and m (N = 192, R2 = 0.979) was found. Moreover, the residuals were normally distributed and the linear regression line was not
signicantly different from the identity line because the condence intervals for the intercept and the slope were equal to 0.01050.0022 and
0.9841.026, respectively.
According to the Er criterion, the vG model yielded, on average, unsatisfactory results. This result, which was obtained locally in a soil having a clay content not exceeding 19.6% (Table 1), was not surprising
because in another investigation carried out in Sicily, the risk of a
weak performance of the vG water retention model used by BEST occurred for soils with a cl content not exceeding 44% (Bagarello and
Iovino, 2012). The performances were generally poorer for lower
values, and this was also an expected result because Haverkamp et al.
(2005) reported that the vG equation provides a relatively poor description of retention data for dry conditions. According to these authors,
using all ve parameters in the optimization may improve data description by Eq. (1a), but this approach was not tested in this investigation
because of the constraints on both r and the relationship between m
and n in the BEST procedure. However, there were signs of a generally
good predictive ability of the model, as denoted by the high R2 value
and the similarity between the vG vs. m regression line and the identity
line. This last result, suggesting that vG was a reasonably good estimate
of the measured value, gave additional justication to the choice to
compare B with vG.
The tted saturated soil water content, s,vG, was lower than at 75%
of the sampling points and the means of s,vG, and were signicantly
different (Table 3). However, the ratio between these means was
equal to 0.97, the regression of against s,vG was statistically signicant
(N = 32, R2 = 0.39), and a coincidence between the vs. s,vG linear regression line and the identity line was plausible, although only according to the 99.5% condence intervals for the intercept and the slope
(Table 3). Therefore, the choice of assuming s = for the BEST calculations was reasonable taking into account that s,vG was not very different
from .
Fig. 1. Relative differences between the volumetric soil water content predicted by tting
the van Genuchten (1980) model to the data, vG (m3 m3), and the measured soil water
content, m (m3 m3), plotted against m (sample size, N = 192).
Statistic
s,vG
Sample size
Mean
Coefcient of variation (%)
Intercept
Slope
Coefcient of determination, R2
99.5% condence interval for the intercept
99.5% condence interval for the slope
32
0.527
5.5
0.207
0.627
0.386
0.0170.431
0.1901.064
32
0.511
5.6
The means were signicantly different at P = 0.05 according to a two-tailed, paired t-test.
The coefcient of correlation, R, was signicantly higher than zero at P = 0.05 according
to a one-tailed t-test.
304
Table 4
Comparison between the experimentally determined soil water content values, vG, and the corresponding estimates obtained with the BEST procedure, B.
Scenario ID
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
28
29
30
31
32
33
34
38
39
40
192
192
192
192
192
192
192
192
192
192
192
192
192
192
192
192
192
192
192
192
186
168
186
186
168
168
168
186
168
186
186
168
168
168
vG (m3 m3)
B (m3 m3)
Me
CV
Me
CV
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.268
0.269
0.269
0.269
0.269
0.269
0.269
0.269
0.269
0.269
0.269
0.269
0.269
0.269
0.269
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
50.5
49.9
50.4
49.9
49.9
50.4
50.4
50.4
49.9
50.4
49.9
49.9
50.4
50.4
50.4
0.242
0.227
0.230
0.231
0.247
0.248
0.249
0.261
0.262
0.262
0.202
0.189
0.190
0.192
0.209
0.210
0.211
0.222
0.223
0.224
0.222
0.198
0.195
0.195
0.249
0.249
0.249
0.182
0.155
0.153
0.153
0.209
0.209
0.209
56.6
58.1
57.9
57.8
55.8
55.7
55.6
53.8
53.7
53.6
72.6
74.6
74.4
74.2
71.6
71.4
71.2
68.7
68.6
68.4
58.2
59.2
59.6
59.6
55.4
55.4
55.4
74.9
76.2
76.6
76.6
71.2
71.2
71.2
b0
b1
R2
Slope
0.022
0.021
0.026
0.025
0.019
0.019
0.018
0.011
0.011
0.010
0.078
0.079
0.079
0.079
0.077
0.077
0.076
0.072
0.071
0.071
0.030
0.028
0.030
0.030
0.019
0.019
0.019
0.081
0.070
0.072
0.072
0.077
0.077
0.077
0.983
0.926
0.954
0.957
0.996
0.997
0.999
1.015
1.015
1.016
1.048
0.998
1.004
1.009
1.066
1.069
1.071
1.098
1.099
1.100
0.935
0.836
0.836
0.836
0.994
0.994
0.994
0.975
0.837
0.837
0.837
1.062
1.062
1.062
0.945
0.902
0.939
0.938
0.950
0.949
0.948
0.956
0.955
0.954
0.928
0.921
0.921
0.920
0.934
0.934
0.933
0.943
0.942
0.942
0.945
0.939
0.937
0.937
0.955
0.955
0.955
0.925
0.921
0.920
0.920
0.936
0.936
0.936
0.032 to 0.012
0.034 to 0.008
0.036 to 0.015
0.036 to 0.015
0.029 to 0.010
0.029 to 0.009
0.028 to 0.008
0.021 to 0.002
0.020 to 0.001
0.019 to 0
0.091 to 0.066
0.091 to 0.066
0.091 to 0.066
0.091 to 0.066
0.089 to 0.065
0.089 to 0.065
0.089 to 0.064
0.083 to 0.060
0.083 to 0.060
0.082 to 0.059
0.039 to 0.020
0.037 to 0.018
0.040 to 0.021
0.040 to 0.021
0.029 to 0.009
0.029 to 0.009
0.029 to 0.009
0.093 to 0.069
0.081 to 0.059
0.083 to 0.061
0.083 to 0.061
0.090 to 0.064
0.090 to 0.064
0.090 to 0.064
0.9491.018
0.8820.969
0.9190.989
0.9220.993
0.9631.029
0.9641.030
0.9651.032
0.9841.046
0.9841.047
0.9841.048
1.0061.089
0.9571.040
0.9611.046
0.9661.052
1.0261.107
1.0281.110
1.0301.112
1.0591.137
1.0601.138
1.0611.139
0.9030.968
0.8030.868
0.8040.868
0.8040.868
0.9601.027
0.9601.027
0.9601.027
0.9351.016
0.7990.874
0.8010.873
0.8010.873
1.0201.105
1.0201.105
1.0201.105
N = number of points included in the comparison; Me = mean value; CV = coefcient of variation; b0, b1 and R2 = intercept, slope and coefcient of determination, respectively, of the
linear regression line of B against vG.
According to a two-tailed, paired t-test, the means of were not signicantly different at P = 0.01 only with reference to scenario no. 10. For each scenario, the correlation coefcient, R,
was signicantly greater than zero at P = 0.05 according to a one-tailed t-test.
identity line (Table 4). The quality of the predictions was also visually
detectable (Fig. 2). Therefore, using the BEST-slope algorithm, the FIT
procedure for shape parameter estimation, and assuming = 1.9 and
= 0.79 yielded predictions of soil water content close to the values obtained experimentally. The Er of the transient inltration model, calculated with Eq. (26) of Lassabatre et al. (2006), was equal to 3.5% on
average and did not exceed 4.6% at an individual sampling point. Taking
into account that an Er 5.5% denotes an acceptable error, the tting
quality was always satisfactory.
0.6
y = 1.0158x 0.0098
R = 0.9541
0.4
B
0.2
data
identy line
0
0
0.2
0.4
0.6
vG
Fig. 2. Regression of the volumetric soil water content predicted with the BEST procedure
of soil hydraulic characterization (scenario no. 10), B (m3 m3), against the soil water
content predicted by tting the van Genuchten (1980) model to the data, vG (m3 m3)
(sample size, N = 192).
The optimal and values among the tested ones were different
from those originally proposed by Lassabatre et al. (2006), equal to
0.6 and 0.75, respectively, but feasible values of these two constants
(Haverkamp et al., 1994; Lassabatre et al., 2009; Nasta et al., 2012)
yielded reliable predictions of in this investigation. An attempt to establish the relative impact of and within their feasible ranges on
the predictions of was therefore carried out by considering scenario
nos. 1 to 10 because a) these scenarios were homogeneous in terms of
algorithm (BEST-slope) and n estimating procedure (FIT), i.e., they
only differed by the values of the two constants, and b) they included
the scenario yielding the best results (scenario no. 10). The mean
value of , Me(), increased with both and (Fig. 3a), but, in general,
Me() was more sensitive to than to . In particular, an increase of
from 0.61 to 0.79 determined an increase of Me() varying from 0.6%
for = 1.9 to 2.1% for = 0.1. On the other hand, an increase of
from 0.1 to 1.9 determined an increase in Me() varying from 13.4%
for = 0.79 to 15.0% for = 0.61. Therefore, the sensitivity of the estimates of Me() to was higher for low values of , and the sensitivity of
these estimates to increased with a decrease in . The choice of the
two inltration constants also affected the estimated relative variability
of , although the coefcients of variation of , CV(), varied in a relatively narrow range, i.e., from 53.6% to 58.1% (Fig. 3b). The CV() values
decreased with an increase in both and , but the former constant affected CV() more than the latter one. According to Nasta et al. (2012),
and have a large impact on the BEST estimation procedure, and both
constants should be calibrated as a function of the soil type. This investigation, carried out on a sandy loam soil, suggested that the choice of
has a larger impact on the estimates of compared with the choice of
within feasible values of these two constants. Therefore, calibration is
particularly important for , i.e., to adjust the relation between the
water diffusivity and the unsaturated hydraulic conductivity (Nasta
et al., 2012).
The * values calculated by Eq. (4) varied from 0.0016 to 0.370 mm1
(N = 32). Excluding two values of * N 0.1 mm1, which were considered unreliable, being expected in gravels and very coarse sands with
negligible amounts of ner soil particles (Reynolds and Lewis, 2012),
the median of * was equal to 0.0067 mm1, and this value was used
305
Fig. 3. Mean, Me (a), and coefcient of variation, CV (b), of the estimated soil water content
for different values of the and constants.
Statistic
BEST
WU
SFH
Minimum
Maximum
Mean
Median
Coefcient of variation (%)
Intercept
Slope
Coefcient of determination, R2
99.5% condence interval for the
intercept
99.5% condence interval for the
slope
0.0023
0.0130
0.0058ab
0.0048
46.2
0.0039
0.0168
0.0067a
0.0058
50.2
0.0008
0.754
0.879
0.00040.0019
0.0170
0.0418
0.0267b
0.0282
25.2
0.0021
0.295
0.546
0.0060.002
0.5990.908
0.1460.443
The means followed by the same letter were compared and they were found to be significantly different at P = 0.05 according to a two-tailed, paired t-test. The coefcients of correlation, R, were signicantly higher than zero at P = 0.05 according to a one-tailed t-test.
306
factor of dicrepancy
10
y = 0.2786x 0.546
R = 0.663
8
6
4
2
0
0
0.005
0.01
BEST esmate of Ks (mm s 1)
0.015
Fig. 4. Comparison between the saturated soil hydraulic conductivity, Ks, values estimated
with the BEST procedure (scenario no. 10) and method 2 by Wu et al. (1999) (WU in the
gure).
Fig. 6. Factor of discrepancy, i.e., ratio between the saturated soil hydraulic conductivity, Ks (mm s 1), obtained with the SFH technique and the BEST procedure, plotted
against the estimate of Ks obtained with BEST.
values because more data were available for than for Ks (168192 versus 32). Moreover, was measured with a standard laboratory technique, whereas method 2 by Wu et al. (1999) and the SFH technique
by Bagarello et al. (2004) cannot be considered standard and commonly
accepted methods for measuring Ks. According to different authors,
standard measurement techniques of this last soil hydraulic property
cannot be identied, and appropriate methods for each specic application should be developed (Bagarello et al., 2014a; Reynolds et al., 2000;
Verbist et al., 2013). Therefore, additional comparisons are clearly advisable in other soils, taking into account that, especially for Ks, the statistical tools alone are not enough to establish the reliability of the BEST
results, but the explicit consideration of what information is contained
in a particular measurement of this variable is also necessary to make
evaluations and judgments. The comparison between the BEST and
the SFH technique was made, taking into account that comparing different techniques for measuring Ks provides one of the few sources of information that practitioners can draw upon to select Ks methods that
are appropriate for their circumstances (Reynolds et al., 2000). The Ks
comparison carried out by applying different procedures to analyze
the same inltration run allowed us to exclude, in the interpretation
of the results, possible effects of spatial variability, which is known to
be particularly noticeable for this soil hydraulic property (Mallants
et al., 1996, 1997; Warrick, 1998). The combined approach used in
this investigation (alternative techniques to determine Ks in the eld,
alternative methods to analyze the inltration run) might also be applied in other circumstances to concretely carry out the conclusion by
Reynolds et al. (2000) that work is still necessary for determining the
most appropriate methods for both the soil conditions and the specic
application of the Ks data.
Fig. 5. Comparison between the saturated soil hydraulic conductivity, Ks, values estimated
with the BEST procedure (scenario no. 10) and the SFH technique (SFH in the gure).
4. Conclusions
In this investigation, the applicability of the BEST procedure for soil
hydraulic characterization was tested in a sandy loam soil. With this
aim, a comparison between the predicted and the laboratory measured
water retention data was established. The saturated soil hydraulic conductivity, Ks, obtained with BEST, method 2 by Wu et al. (1999) applied
to the BEST inltration run and the Simplied Falling Head (SFH) technique, was also compared.
The water retention model used by BEST satisfactorily reproduced
the experimental data, and the choice to assume a saturated soil water
content coinciding with soil porosity, frequently made in several BEST
applications, was reasonable because the two parameters were not
very different for the investigated soil.
A good correspondence between the predicted and the experimental water retention data was detected when the original BEST-slope algorithm was applied with the inltration constants set at = 1.9 and
= 0.79. In this case, BEST also yielded plausible Ks values in that
they differed by not more than a factor of two from the corresponding
307
Di Prima, S., 2013. Automatic analysis of multiple Beerkan inltration experiments for soil
hydraulic characterization. CIGR Inter-Regional Conference on Land and Water
Challenges Bari (Italy), 1014 September, 2013.
Elrick, D.E., Reynolds, W.D., 1992. Methods for analyzing constant-head well permeameter
data. Soil Sci. Soc. Am. J. 56, 320323.
Gee, G.W., Bauder, J.W., 1986. Particle-size analysis, In: Klute, A. (Ed.), Methods of Soil
Analysis, Part 1: Physical and Mineralogical Methods, 2nd ed. Agron. Monogr., 9.
ASA and SSSA, Madison, WI, pp. 383411.
Gonzalez-Sosa, E., Braud, I., Dehotin, J., Lassabatre, L., Angulo-Jaramillo, R., Lagouy, M.,
Branger, F., Jacqueminet, C., Kermadi, S., Michel, M., 2010. Impact of land use on the
hydraulic properties of the topsoil in a small French catchment. Hydrol. Process. 24,
23822399.
Haverkamp, R., Ross, P.J., Smettem, K.R.J., Parlange, J.Y., 1994. Three-dimensional analysis
of inltration from the disc inltrometer: 2. Physically based inltration equation.
Water Resour. Res. 30, 29312935.
Haverkamp, R.,Leij, F.J.,Fuentes, C., Sciortino, A., Ross, P.J., 2005. Soil water retention: I. Introduction of a shape index. Soil Sci. Soc. Am. J. 69, 18811890.
Helsel, D.R., Hirsch, R.M., 1992. Statistical Methods in Water Resources. Elsevier Science
Publishers, Amsterdam, The Netherlands 0-444-81463-9 (522 pp.).
Lai, J.,Luo, Y.,Ren, L., 2010. Buffer index effects on hydraulic conductivity measurements using
numerical simulations of double-ring inltration. Soil Sci. Soc. Am. J. 74, 15261536.
Lassabatre, L., Angulo-Jaramillo, R., Soria Ugalde, J.M., Cuenca, R., Braud, I.,
Haverkamp, R., 2006. Beerkan estimation of soil transfer parameters through inltration experiments BEST. Soil Sci. Soc. Am. J. 70, 521532.
Lassabatre, L., Angulo-Jaramillo, R., Soria, Ugalde J., imnek, J., Haverkamp, R., 2009. Numerical evaluation of a set of analytical inltration equations. Water Resour. Res. 45.
http://dx.doi.org/10.1029/2009WR007941 (W12415).
Lassabatre, L., Angulo-Jaramillo, R., Goutaland, D., Letellier, L., Gaudet, J.P., Winiarski, T.,
Delolme, C., 2010. Effect of settlement of sediments on water inltration in two
urban inltration basins. Geoderma 156, 316325.
Lilliefors, H.W., 1967. On the KolmogorovSmirnov test for normality with mean and variance unknown. J. Am. Stat. Assoc. 62 (318), 399402.
Mallants, D.,Mohanty, B.P.,Jacques, D.,Feyen, J., 1996. Spatial variability of hydraulic properties in a multi-layered soil prole. Soil Sci. 161 (3), 167181.
Mallants, D., Mohanty, B.P., Vervoort, A., Feyen, J., 1997. Spatial analysis of saturated hydraulic conductivity in a soil with macropores. Soil Technol. 10, 115131.
Minasny, B.,McBratney, A.B., 2007. Estimating the water retention shape parameter from
sand and clay content. Soil Sci. Soc. Am. J. 71, 11051110.
Mubarak, I., Mailhol, J.C., Angulo-Jaramillo, R., Ruelle, P., Boivin, P., Khaledian, M., 2009a.
Temporal variability in soil hydraulic properties under drip irrigation. Geoderma
150, 158165.
Mubarak, I.,Mailhol, J.C.,Angulo-Jaramillo, R.,Bouarfa, S.,Ruelle, P., 2009b. Effect of temporal variability in soil hydraulic properties on simulated water transfer under highfrequency drip irrigation. Agric. Water Manag. 96, 15471559.
Mubarak, I., Angulo-Jaramillo, R., Mailhol, J.C., Ruelle, P., Khaledian, M., Vauclin, M., 2010.
Spatial analysis of soil surface hydraulic properties: is inltration method dependent?
Agric. Water Manag. 97, 15171526.
Nasta, P., Lassabatere, L., Kandelous, M.M., imnek, J., Angulo-Jaramillo, R., 2012. Analysis
of the role of tortuosity and inltration constants in the Beerkan method. Soil Sci. Soc.
Am. J. http://dx.doi.org/10.2136/sssaj/2012.0117n.
Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon and organic matter. In:
Sparks, D.L., et al. (Eds.), Methods of Soil Analysis, Part 3, Chemical Method. SSSA
Book Series, 5. Soil Science Society of America, Inc., Madison, WI, pp. 9611010.
Reynolds, W.D., Elrick, D.E., 2005. Chapter 6 Measurement and characterization of soil hydraulic properties, in: lvarez-Bened, J., Muoz-Carpena, R. (co-eds.), SoilWater
Solute Process Characterization An Integrated Approach. CRC Press, Boca Raton,
Florida, USA, p. 197252.
Reynolds, W.D., Lewis, J.K., 2012. A drive point application of the Guelph Permeameter
method for coarse-textured soils. Geoderma 187188, 5966.
Reynolds, W.D.,Bowman, B.T., Brunke, R.R., Drury, C.F., Tan, C.S., 2000. Comparison of tension inltrometer, pressure inltrometer, and soil core estimates of saturated hydraulic conductivity. Soil Sci. Soc. Am. J. 64, 478484.
van Genuchten, M.Th, 1980. A closed form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44, 892898.
Verbist, K.M.J., Cornelis, W.M., Torfs, S., Gabriels, D., 2013. Comparing methods to determine hydraulic conductivities on stony soils. Soil Sci. Soc. Am. J. 77 (1), 2542.
http://dx.doi.org/10.2136/sssaj2012.0025.
Warrick, A.W., 1998. Spatial variability. In: Hillel, D. (Ed.), Environmental Soil Physics. Academic Press, San Diego, pp. 655675 (771 pp.).
Wu, L., Pan, L., 1997. A generalized solution to inltration from single-ring inltrometers
by scaling. Soil Sci. Soc. Am. J. 61, 13181322.
Wu, L.,Pan, L.,Mitchell, J.,Sanden, B., 1999. Measuring saturated hydraulic conductivity using
a generalized solution for single-ring inltrometers. Soil Sci. Soc. Am. J. 63, 788792.
Wuest, S.B., 2005. Bias in ponded inltration estimates due to sample volume and shape.
Vadose Zone J. 4, 11831190.
Xu, X., Kiely, G., Lewis, G., 2009. Estimation and analysis of soil hydraulic properties
through inltration experiments: comparison of BEST and DL tting methods. Soil
Use Manag. 25, 354361.
Xu, X.,Lewis, C.,Liu, W.,Albertson, J.D.,Kiely, G., 2012. Analysis of single-ring inltrometer
data for soil hydraulic properties estimation: comparison of BEST and Wu methods.
Agric. Water Manag. 107, 3441.
Yilmaz, D.,Lassabatre, L.,Angulo-Jaramillo, R.,Deneele, D.,Legret, M., 2010. Hydrodynamic
characterization of basic oxygen furnace slag through an adapted BEST method.
Vadose Zone J. 9, 110.
Youngs, E.G., 1987. Estimating hydraulic conductivity values from ring inltrometer measurements. J. Soil Sci. 38, 623632.